Abstract
To understand the relative influences of operational taxonomic units (OTUs) and amplicon sequence variants (ASVs) on patterns of marine microbial diversity and community composition, we examined bacterial diversity and community composition of seawater from 12 sites in the North Atlantic Ocean and Canadian Arctic and sediment from two sites in the North Atlantic. For the seawater analyses, we included samples from three to six zones in the water column of each site. For the sediment analyses, we included over 20 sediment horizons at each of two sites. For all samples, we amplified the V4–V5 hypervariable region of the 16S ribosomal RNA (rRNA) gene. We analyzed each sample in two different ways: (i) by clustering its reads into 97%-similar OTUs and (ii) by assigning sequences to unique ASVs. OTU richness is much higher than ASV richness for every sample, but both OTUs and ASVs exhibit similar vertical patterns of relative diversity in both the water column and the sediment. Bacterial richness is highest just below the photic zone in the water column and at the seafloor in the sediment. For both OTUs and ASVs, richness estimates depend on the number of sequences analyzed. Both methods yield broadly similar community compositions for each sample at the taxonomic levels of phyla to families. While the two methods yield different richness values, broad-scale patterns of relative richness and community composition are similar with both methods.
Highlights
Some studies have relied on operational taxonomic units (OTUs; Huse et al, 2010; Jing et al, 2013; Bienhold et al, 2016; Sheik et al, 2018; Kirkpatrick et al, 2019; Pohlner et al, 2019), whereas others have relied on amplicon sequence variants (ASVs; Callahan et al, 2017; Hoshino et al, 2020; Fadeev et al, 2021)
Eighteen thousand six hundred ninety-six bacterial ASVs were generated from all samples from a total of 2,402,375 sequences with the DADA2 pipeline (Supplementary Tables S3 and S4)
While many studies in recent years have highlighted the differences between bioinformatic pipelines, our comparison of OTU-based results and ASV-based results demonstrates that both methods yield similar patterns of relative taxonomic richness and evenness in seawater and marine sediment
Summary
High-throughput sequencing (HTS) has dominated studies of microbial diversity and community composition in marine environments (Inagaki et al, 2006; Bienhold et al, 2016; Kirkpatrick et al, 2019) and other environments (He et al, 2015; Tremblay et al, 2015; Berry et al, 2017; Glassman and Martiny, 2018). Studies that rely on OTUs typically cluster sequences at an operationally defined level of similarity (usually 97%; Ward et al, 1998; Roesch et al, 2007; Huse et al, 2010; He et al, 2015; Mysara et al, 2017) This clustering is intended to reduce the impact on diversity estimates of errors that accumulate during PCR amplification and genetic sequencing (Huse et al, 2010; Kunin et al, 2010). Studies that rely on ASVs use algorithms intended to remove the errors associated with sequencing and return individual, unique sequences that represent individual taxa (Rosen et al, 2012; Eren et al, 2013; Callahan et al, 2016; Edgar, 2016). These sequences (ASVs or ESVs—exact sequence variants) differ from each other by as little as a single nucleotide change (Rosen et al, 2012; Callahan et al, 2016; Edgar, 2016)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
